In a time division multiple access (TDMA) cellular telephone system, an analog voice signal is delivered to the base station of a cell for transmission to a remote user or mobile station in the cell by means of a radio frequency (RF) downlink signal. At the base station, the analog voice signal is first digitized. The digitized voice signal is next compressed using known voice compression techniques. In order to preserve the quality of the signal during transmission to the mobile station, forward error protection data is added to the compressed voice signal. Forward error protection is a known signal processing technique that allows the mobile station to recover valid data in the presence of transmission errors. The compressed voice signal with forward error protection data is then multiplexed with other compressed voice signals having forward error protection data and transmitted as an RF signal to the mobile stations within the cell serviced by the base station. The transmitted compressed voice signal with forward error protection data is received by the mobile station, decompressed, and converted to an analog signal to recover the original voice signal. In the same fashion, the mobile station may also digitize, compress, add forward error protection, and transmit the compressed voice signal with forward error protection back to the base station.
The voice signal compression or encoding process at the base station is done using known voice encoders (vocoders) 15 and data compression techniques. Likewise, the decompression process at the mobile station or remote user is done using known decoders and known decompression techniques. The voice signal is compressed and decompressed in order to conserve bandwidth within the RF transmission spectrum. Adding forward error protection data to the compressed voice signal requires additional bandwidth.
The amount of signal compression of the vocoder is quantified by the ratio of the input data rate of the digitized voice signal to the output data rate for the compressed voice signal. For instance, if the digitized voice signal input to the vocoder is 64 kilobits per second (kbps) and if the output from the vocoder is 8 kbps, then the vocoder has compressed the voice signal 8 times and has an 8:1 compression ratio. The capacity of a digital TDMA cellular telephone system is determined by the bit rate needed for each mobile station to communicate versus the total bit rate that the base station can support. For example, if each mobile station needs 10 kbps of bandwidth and if the base station can support 100 kbps of band width, then the base station can support 10 mobile stations.
In a TDMA cellular telephone system (such as specified by TIA Standard IS-136, which is incorporated herein by reference), the RF transmission spectrum is divided up into smaller portions of spectrum, called channels, which in turn are then time shared by a number of mobile stations. For instance, TIA Standard IS-136 defines a base station for a TDMA cellular telephone system which has a set of RF channels, each 30 kilohertz (kHz) in bandwidth. Each RF channel is time divided into frames, and each frame is divided into 6 equally spaced time slots as shown in FIG. 3. The length of each frame is 40 milliseconds or 1944 bits or 972 symbols. The length of each time slot is 6.67 milliseconds or 324 bits or 162 symbols.
The mobile stations served by a single 30 kHz RF channel are assigned different time slots so that the mobile stations can share the RF channel by communicating (transmitting or receiving a voice signal) only within the prescribed time slots of the RF channel. TIA Standard IS-136 defines two types of voice transmission, half rate in which each mobile station only uses one time slot out of the 6 per frame and full rate in which each mobile station uses two time slots out of 6 per frame. Therefore, the number of voice channels that are available for each 30 kHz RF channel is 6 for half rate and 3 for full rate operation.
The quality of the voice signal received by a mobile station is dependent on the degree of signal compression, the amount of forward error protection data transmitted, and the strength of the RF signal at the mobile station's location. Generally, vocoders that compress speech to a lower bit rate (higher compression ratio) will have a lower voice quality than vocoders with a higher bit rate output (lower compression ratio). Also, the quality of the signal received by the mobile station is improved by transmitting more forward error protection data with the compressed voice signal. Both high bit rate and added forward error protection data require additional transmission bandwidth. Therefore, there is a direct tradeoff between the voice quality that a mobile station will experience versus the capacity of the base station.
The quality of the voice signal received by a mobile station is also effected by the strength of the RF signal at the mobile station's location. As the mobile stations move away from the base station, the strength of the RF signal diminishes, and the quality of the voice transmission may deteriorate as a result.